Aditi Sharan

Keyword and Keyphrase Extraction Techniques: A Literature Review

In this paper we present a survey of various techniques available in text mining for keyword and keyphrase extraction.

An improved intuitionistic fuzzy c-means clustering algorithm incorporating local information for brain image segmentation

Original and segmented simulated brain image by different algorithms: (a) axial view of original simulated T1-weighted brain image with INU=0 and 1% noise, (b) skull stripping simulated brain image, (c) manual segmented CSF, GM and WM images, (d) IIFCM algorithm, (e) IFCM algorithm, (f) FLICM algorithm, (g) EnFCM algorithm, (h) FGFCM algorithm, (i) FCM_S1 algorithm, (j) FCM_S2 algorithm, (k) ImFCM algorithm. The segmentation of brain magnetic resonance (MR) images plays an important role in the computer-aided diagnosis and clinical research. However, due to presence of noise and uncertainty on the boundary between different tissues in the brain image, the segmentation of brain image is a challenging task. Many variants of standard fuzzy c-means (FCM) algorithm have been proposed to handle the noise. Intuitionistic fuzzy c-means (IFCM) algorithm, one of the variants of FCM, is found suitable for image segmentation. It incorporates the advantage of intuitionistic fuzzy sets theory. The IFCM successfully handles the uncertainty but it is sensitive to noise as it does not incorporate any local spatial information. In this paper, we have presented a novel approach, named an improved intuitionistic fuzzy c-means (IIFCM), which considers the local spatial information in an intuitionistic fuzzy way. The IIFCM preserves the image details, is insensitive to noise, and is free of requirement of any parameter tuning. The obtained segmentation results on synthetic square image, real and simulated MRI brain image demonstrate the efficacy of the IIFCM algorithm and superior performance in comparison to existing segmentation methods. A nonparametric statistical analysis is also carried out to show the significant performance of the IIFCM algorithm in comparison to other existing segmentation algorithms.

Personalized web search using browsing history and domain knowledge

Generic search engines are important for retrieving relevant information from web. However these engines follow the “one size fits all” model which is not adaptable to individual users. Personalized web search is an important field for tuning the traditional IR system for focused information retrieval. This paper is an attempt to improve personalized web search. Users Profile provides an important input for performing personalized web search. This paper proposes a framework for constructing an Enhanced User Profile by using users browsing history and enriching it using domain knowledge. This Enhanced User Profile can be used for improving the performance of personalized web search. In this paper we have used the Enhanced User Profile specifically for suggesting relevant pages to the user. The experimental results show that the suggestions provided to the user using Enhanced User Profile are better than those obtained by using a User Profile.

Relevance feedback based query expansion model using Borda count and semantic similarity approach

Pseudo-Relevance Feedback (PRF) is a well-known method of query expansion for improving the performance of information retrieval systems. All the terms of PRF documents are not important for expanding the user query. Therefore selection of proper expansion term is very important for improving system performance. Individual query expansion terms selection methods have been widely investigated for improving its performance. Every individual expansion term selection method has its own weaknesses and strengths. To overcome the weaknesses and to utilize the strengths of the individual method, we used multiple terms selection methods together. In this paper, first the possibility of improving the overall performance using individual query expansion terms selection methods has been explored. Second, Borda count rank aggregation approach is used for combining multiple query expansion terms selection methods. Third, the semantic similarity approach is used to select semantically similar terms with the query after applying Borda count ranks combining approach. Our experimental results demonstrated that our proposed approaches achieved a significant improvement over individual terms selection method and related state-of-the-art methods.

Context Window Based Co-occurrence Approach for Improving Feedback Based Query Expansion in Information Retrieval

Pseudo-relevance feedback (PRF) is a type of relevance feedback approach of query expansion that considers the top ranked retrieved documents as relevance feedback. In this paper the authors focus is to capture the limitation of co-occurrence and PRF based query expansion approach and the authors proposed a hybrid method to improve the performance of PRF based query expansion by combining query term co-occurrence and query terms contextual information based on corpus of top retrieved feedback documents in first pass. Firstly, the paper suggests top retrieved feedback documents based query term co-occurrence approach to select an optimal combination of query terms from a pool of terms obtained using PRF based query expansion. Second, contextual window based approach is used to select the query context related terms from top feedback documents. Third, comparisons were made among baseline, co-occurrence and contextual window based approaches using different performance evaluating metrics. The experiments were performed on benchmark data and the results show significant improvement over baseline approach.

Co-occurrence based predictors for estimating query difficulty

Query difficulty prediction aims to identify, in advance, how reliably an information retrieval system will perform when faced with a particular user request. The prediction of query difficulty level is an interesting and important issue in Information Retrieval (IR) and is still an open research. In order to appreciate importance of query difficulty prediction we present an example., Information Retrieval (IR) is the Science of searching the relevant documents based on user’s need and a way towards discovering knowledge from text data. User’s needs are often expressed in terms of query. It has been observed that there is a word mismatch problem while matching user’s query to the documents. This is because users and authors of documents do not use same vocabulary. Query expansion/reformulation is a method to overcome such mismatch in terminology. Query expansion (QE) has become a well known technique that has been shown to improve average retrieval performance. However despite extensive research QE does not provide consistent gains over different query sets and collections. Therefore this technique has not been used in many operational systems as it may degrade performance of individual queries. A thorough investigation into robustness of query expansion is required in order to ensure reliability of query expansion for individual queries. It is well-known in the Information Retrieval community that methods such as query expansion can help ”easy” queries but are detrimental to ”hard” queries If the performance of queries can be predicted before retrieval then specific measures can be taken to improve the overall performance of the system. In this paper we do thorough investigations of various query difficulty predictors l and suggest two new query predictorsl based on co-occurrence of query terms. To evaluate the predictors, we have experimented on standard TREC collections. Our work is significant as it is a step towards judging reliability and robustness of query processing operations such as query expansion.

A new fuzzy logic-based query expansion model for efficient information retrieval using relevance feedback approach

Abstract Efficient query expansion (QE) terms selection methods are really very important for improving the accuracy and efficiency of the system by removing the irrelevant and redundant terms from the top-retrieved feedback documents corpus with respect to a user query. Each individual QE term selection method has its weaknesses and strengths. To overcome the weaknesses and to utilize the strengths of the individual method, we used multiple terms selection methods together. In this paper, we present a new method for QE based on fuzzy logic considering the top-retrieved document as relevance feedback documents for mining additional QE terms. Different QE terms selection methods calculate the degrees of importance of all unique terms of top-retrieved documents collection for mining additional expansion terms. These methods give different relevance scores for each term. The proposed method combines different weights of each term by using fuzzy rules to infer the weights of the additional query terms. Then, the weights of the additional query terms and the weights of the original query terms are used to form the new query vector, and we use this new query vector to retrieve documents. All the experiments are performed on TREC and FIRE benchmark datasets. The proposed QE method increases the precision rates and the recall rates of information retrieval systems for dealing with document retrieval. It gets a significant higher average recall rate, average precision rate and F measure on both datasets.

Co-occurrence and Semantic Similarity Based Hybrid Approach for Improving Automatic Query Expansion in Information Retrieval

Pseudo Relevance feedback PRF based query expansion approaches assumes that the top ranked retrieved documents are relevant. But this assumption is not always true; it may also possible that a PRF document may contain different topics, which may or may not be relevant to the query terms even if the documents are judged relevant. In this paper our focus is to capture the limitation of PRF based query expansion and propose a hybrid method to improve the performance of PRF based query expansion by combining corpus based term co-occurrence information and semantic information of term. Firstly, the paper suggest use of corpus based term co-occurrence approach to select an optimal combination of query terms from a pool of terms obtained using PRF based query expansion. Second, we use semantic similarity approach to rank the query expansion terms obtained from top feedback documents. Third, we combine co-occurrence and semantic similarity together to rank the query expansion terms obtained from first step on the basis of semantic similarity. The experiments were performed on FIRE ad hoc and TREC-3 benchmark datasets of information retrieval. The results show significant improvement in terms of precision, recall and mean average precision MAP. This experiments shows that the combination of both techniques in an intelligent way gives us goodness of both of them. As this is the first attempt in this direction there is a large scope of improving these techniques.

Hybrid Approach for Single Text Document Summarization Using Statistical and Sentiment Features

Summarization is a way to represent same information in concise way with equal sense. This can be categorized in two type Abstractive and Extractive type. Our work is focused around Extractive summarization. A generic approach to extractive summarization is to consider sentence as an entity, score each sentence based on some indicative features to ascertain the quality of sentence for inclusion in summary. Sort the sentences on the score and consider top n sentences for summarization. Mostly statistical features have been used for scoring the sentences. We are proposing a hybrid model for a single text document summarization. This hybrid model is an extraction based approach, which is combination of Statistical and semantic technique. The hybrid model depends on the linear combination of statistical measures : sentence position, TF-IDF, Aggregate similarity, centroid, and semantic measure. Our idea to include sentiment analysis for salient sentence extraction is derived from the concept that emotion plays an important role in communication to effectively convey any message hence, it can play a vital role in text document summarization. For comparison we have generated five system summaries Proposed Work, MEAD system, Microsoft system, OPINOSIS system, and Human generated summary, and evaluation is done using ROUGE score.

A framework for restricted domain Question Answering System

This paper proposes a framework for developing Question Answering System for restricted domain using advanced NLP tools. The proposed model basically works over the concept of Information Extraction rather than the old technique of information Retrieval used by the search engines. The main objective of the model is to extract the exact and precise answer for the given question from a large dataset. This framework is simple and easy to implement against the previously developed complex architectures. The Framework is divided into four modules namely: Question Processing Module, Document Processing Module, Paragraph extraction module and Answer extraction module. The paper also proposes various algorithms separately for Definition Type, Descriptive Type and Factoid Type of questions for extracting most potential answer from the large dataset.